Susceptibility to Influenza Virus Infection of Bronchial Biopsies in Asthma

Susceptibility to influenza virus infection of bronchial biopsies in asthma

Ben Nicholas, PhDa,b, Sarah Dudley, PhDc, Kamran Tariq, MDa,b, Peter Howarth, DMa,b, Kerry Lunn, BScc, Sandy Pink, BAb, Peter J. Sterk, PhDd, Ian M. Adcock, PhDe, Phillip Monk, PhDc, Ratko Djukanovic, MDa,b, On behalf of the U-BIOPRED study group

aClinical and Experimental Sciences, University of Southampton Faculty of Medicine, Sir Henry Wellcome Laboratories, Southampton General Hospital,

bSouthampton NIHR Respiratory Biomedical Research Unit and the NIHR Wellcome Trust Clinical Research Facility, Southampton General Hospital,

cSynairgen Research Ltd, Southampton General Hospital, Southampton, United Kingdom;

dDepartment of Respiratory Medicine, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands;

eFaculty of Medicine, National Heart and Lung Institute, Imperial College, London, United Kingdom.

Author for Correspondence:

Dr Ben Nicholas,

Clinical and Experimental Sciences,

University of Southampton Faculty of Medicine,

Sir Henry Wellcome Laboratories,

Southampton General Hospital,

E-mail: .

Influenza causes significant morbidity and mortality, especially in patients with chronic lung diseases.1 Infection results in inflammatory cell influx and leads to either resolution or increased lung immunopathology and resulting morbidity,2 especially in patients with chronic airways diseases where viruses exacerbate inflammation and, subsequently, symptoms.

Those with asthma are more susceptible to influenza and are, therefore, the most common population hospitalized, although, interestingly, they are less likely to develop severe disease or die than those without asthma.3 The mechanisms underlying the increased susceptibility to viral infections in those with asthma are poorly understood, but it has been suggested that the skewing toward TH2 immunity results in deficient TH1 antiviral immunity.4 Understanding of antiviral immunity in asthma is also complicated by the immunosuppressive effects of inhaled corticosteroids (ICSs) or oral corticosteroids, standard treatments in asthma.5 The effectiveness of ICSs during exacerbations is unclear because doubling their dose at the time of upper respiratory tract infection fails to prevent asthma exacerbations.6 Corticosteroids may protect against severe outcomes in those with asthma with influenza infection, whereas systemic corticosteroids in individuals without asthma cause delayed viral clearance.7

The primary aim of our study was to compare the susceptibility and inflammatory responses to influenza virus infection of ICS treated patients with asthma and healthy individuals. We hypothesized that these patients with asthma are more susceptible to influenza infection and that their inflammatory responses during infection are elevated, thereby contributing to asthma exacerbations. In view of ethical and safety difficulties of studying influenza infection in vivo, especially in patients with asthma, the airway responses were studied in bronchial biopsies infected ex vivo, using a bronchial biopsy explant model.8 To mimic in vivo conditions of viral exposure, bronchial biopsies from patients with asthma regularly treated with ICSs were exposed to influenza virus in the presence of exogenous ICS, fluticasone propionate (FP), whereas biopsies from healthy subjects were infected in the absence of this corticosteroid.

Twenty-four hours after ex vivo infection, biopsies were enzymatically dispersed with collagenase, allowing quantification of infected cells and activation markers by multicolor flow cytometry (see details in this article’s Online Repository at In these conditions, epithelial cell infection was not different between health and asthma (Fig 1, Ai), whereas viral shedding was significantly reduced in explants from patients with asthma (Fig 1, Aii). T-lymphocyte activation induced by infection (measured by fold-induction of cell surface HLA-DR expression) was suppressed in the biopsies from patients with asthma (Fig 1, Bi); HLA-DR expression on epithelial cells was unchanged (Fig 1, Bii). Secreted mediator responses, including innate defence (IFN-, C-X-C motif chemokine 10 [CXCL-10]), chemokines (CXCL-8, monocyte chemoattractant protein 1, macrophage inflammatory protein 1), and proinflammatory cytokines (IL-1, IL-6, TNF-α) were also all blunted in those with asthma when compared with healthy controls (Fig 1, C). Type I interferons were not present in sufficient quantity to measure; however, the finding of lower CXCL-10 quantities in asthmatic explants was consistent with deficient innate antiviral defences in asthma.

Because subjects with asthma in our study were on regular ICSs, the effects on influenza susceptibility of which are unknown, we also sought to determine whether the differences seen in the primary comparator groups reflected the effects of FP or disease. Ex vivo treatment of steroid-naive bronchial explants from healthy participants with FP increased the percentage of virally infected epithelial cells (Fig 2, Ai) without (in contrast to asthmatic explants) affecting viral shedding (Fig 2, Aii) or activation of T lymphocytes (Fig 2, Bi), but epithelial cell surface induction of HLA-DR was suppressed (Fig 2, Bii). We have previously observed elevation of this panel ofmediators with influenza infection in healthy subjects, with the exception of IL-8.8 As expected, FP treatment significantly inhibited the secretion of these mediators (Fig 2, C).

This study points to important differences between asthma and health in respect of influenza virus handling. However, our hypothesis that the elevated morbidity and mortality caused by influenza infection in people with asthma is associated with increased susceptibility to infection was not fully supported because the initial infection rate was no different between asthma and health (judged by similar proportions of infected epithelial cells). Nevertheless, the blunted inflammatory, including innate immune, responses and T-cell activation (judged by lesser induction of cell surface HLA-DR expression) to infection in ICS-treated patients with asthma do argue in favor of deficient anti-influenza immunity in these patients. Although the reduced viral shedding in asthmatic tissues (when compared with healthy tissues) could be viewed as a positive phenomenon that limits virus spread, alternatively, it could account for prolonged viral retention, with consequential prolonged recovery and increased risk of viremia. Prolonged viral shedding appears to be a consistent problem associated with systemic corticosteroid therapy in patients hospitalized with influenza, in contrast to antiviral agents that enhance virus clearance.7

For ethical and safety reasons, it was impossible to wash out the effects of regular treatment with ICS on the asthmatic explant responses to infection. Accepting that the effects of corticosteroids likely differ between healthy and asthmatic tissue, we thought it useful to study the impact of FP treatment on explants from healthy subjects. In contrast to asthma, this showed that FP increased epithelial infection rates, while viral shedding and T-cell activation were unaffected. Similar to asthma, mediator secretion was suppressed. We also found that MHC class II (HLA-DR) induction in healthy airway tissue epithelial cells was suppressed by ex vivo FP treatment. We have previously observed influenza infection-mediated elevation in HLA-DR on the surface of primary bronchial epithelial cells,9 an effect replicated in bronchial epithelial cells of our explants, probably occurring in both models as a result of infection-induced secretion of IFN-, and suggestive that respiratory epithelial cells potentiate cytotoxic T-cell activity.

Corticosteroid suppression of this effect may reflect suppression of innate antiviral mediators including IFN-, and could have implications for clearance of virally infected cells from the lungs. In summary, the present study shows blunted responses of ICS treated patients with mild/moderate asthma to influenza virus infection but is unable to differentiate between the impact of corticosteroid and disease itself. The lack of effect of corticosteroids in explants from healthy participants suggests that reduced viral shedding and defective T-cell activation observed in patients with asthma may be independent of corticosteroid treatment. Further study is needed to elucidate the underlying mechanisms.

Acknowledgements

The Unbiased BIOmarkers Predictive of REspiratory Disease outcomes project is supported through an Innovative Medicines Initiative Joint Undertaking under grant agreement no. 115010.

Disclosure of potential conflict of interest: S. Dudley and K. Lunn are both employed by and have received share options from Synairgen Research Ltd. P. Howarth’s institution has received travel support from European Union (EU)/Innovative Medicines Initiative (IMI); he has personally received board membership from Roche, Novartis, and Boehringer-Ingelheim; is employed by GlaxoSmithKline (GSK); has received payment for lectures from GSK and Novartis; and has stock options with GSK. P. J. Sterk’s institution has received a public-private grant by the IMI paid by the EU and the European Federation of Pharmaceutical Industries and Associations (EFPIA) for the Unbiased BIOmarkers Predictive of REspiratory Disease outcomes (U-BIOPRED) project. I. M. Adcock’s institution has received a grant from U-BIOPRED Severe asthma for this work; has received a grant from Wellcome Trust; and has personally received board membership from Chiesi; speaker fees from GSK and AZ; and travel assistance from Boehringer-Ingelheim. P. Monk’s institution has received grant no. 115010 from EU’s Seventh Framework Programme (FP7/2007-2013) and EFPIA companies’ in kind contribution. P. Monk is employed by and holds shares of Synairgen Research Ltd. R. Djukanovic has given lectures at symposiums organized by pharmaceutical companies (eg, Novartis and TEVA); consulted at company as a member of advisory boards (eg, Novartis and TEVA); and is a cofounder, current consultant, and shareholder in Synairgen. The rest of the authors declare that they have no relevant conflicts of interest.

REFERENCES

1. Gerke AK, Yang M, Tang F, Foster ED, Cavanaugh JE, Polgreen PM. Association of hospitalizations for asthma with seasonal and pandemic influenza. Respirology 2014;19:116-21.

2. La Gruta NL, Kedzierska K, Stambas J, Doherty PC. A question of selfpreservation: immunopathology in influenza virus infection. Immunol Cell Biol 2007;85:85-92.

3. Van Kerkhove MD, Vandemaele KA, Shinde V, Jaramillo-Gutierrez G, Koukounari A, Donnelly CA, et al. WHO Working Group for Risk Factors for Severe H1N1pdm Infection. Risk factors for severe outcomes following 2009 influenza A (H1N1) infection: a global pooled analysis. PLoS Med 2011;8:e1001053.

4. Message SD, Laza-Stanca V, Mallia P, Parker HL, Zhu J, Kebadze T, et al. Rhinovirus- induced lower respiratory illness is increased in asthma and related to virus load and Th1/2 cytokine and IL-10 production. Proc Natl Acad Sci U S A 2008; 105:13562-7.

5. Myles P, Nguyen-Van-Tam JS, Semple MG, Brett SJ, Bannister B, Read RC, et al, Influenza Clinical Information Network FLU-CIN. Differences between asthmatics and nonasthmatics hospitalised with influenza A infection. Eur Respir J 2013;41: 824-31.

6. Harrison TW, Oborne J, Newton S, Tattersfield AE. Doubling the dose of inhaled corticosteroid to prevent asthma exacerbations: randomised controlled trial. Lancet 2004;363:271-5.

7. Lee N, Chan PK, Hui DS, Rainer TH, Wong E, Choi KW, et al. Viral loads and duration of viral shedding in adult patients hospitalized with influenza. J Infect Dis 2009;200:492-500.

8. Nicholas B, Staples KJ, Moese S, Meldrum E, Ward J, Dennison P, et al. A novel lung explant model for the ex vivo study of efficacy and mechanisms of antiinfluenza drugs. J Immunol 2015;194:6144-54.

9. Wilkinson TM, Li CK, Chui CS, Huang AK, Perkins M, Liebner JC, et al. Nat Med 2012;18:274-80.

Figure Legends

Figure 1. Comparison of bronchial biopsies from healthy subjects and subjects with asthma following influenza virusexposure. A, Effectson (i) epithelial cell infectionand(ii) viral shedding. B, Fold change inMHCclass II cell surface expression on (i) T lymphocytes and (ii) epithelial cells. C, Mediator secretion from infected biopsies. N510 per group compared using Mann-Whitney test.MCP, Monocyte chemoattractant protein; MIP, macrophage inflammatory protein; NP, Influenza A virus nucleoprotein. *P < .05, **P < .005, and ***P < .001.

Figure 2. A, Effect of addition of exogenous corticosteroid to bronchial explants from healthy subjects on (i) epithelial cell infection and (ii) viral shedding. B, Fold change in MHC class II cell surface expression following infection on the cell surface of (i) T lymphocytes and (ii) epithelial cells. C, Mediator secretion from infected biopsies. N 5 10 per group. Data were compared using Wilcoxon matched pairs signed rank test. MCP, Monocyte chemoattractant protein; MIP, macrophage inflammatory protein; NP, Influenza A virus nucleoprotein. *P < .05, **P < .005, and ***P < .001.

Figure 1

Figure 2

Supplementary Files

METHODS

Influenza virus preparation A/H3N2/Wisconsin/67/2005 seed stocks were obtained from the National Institute for Biological Standards and Control, propagated in embryonated specific pathogen free chicken eggs, and, subsequently, purified from egg allantoic fluid by sucrose density gradient ultracentrifugation (Virapur LLC, San Diego, Calif). Stock and conditioned media viral titers were determined by MDCK plaque assay (see details below).

Participants and sample collection Ten healthy participants and 10 subjects with mild/moderate asthma on regular ICSs were recruited as part of the Unbiased BIOmarkers Predictive of REspiratory Disease outcomes project.E1 Participants were matched for age and sex, but subjects with asthma had significantly reduced lung function and evidence of increased airway inflammation and atopy (Table E1). Fiberoptic bronchoscopy was performed according to a standard research protocol.E2 Up to 10 endobronchial biopsies were taken from the subcarinae of large airways.

Ex vivo infection of bronchial explants

The protocol for ex vivo infection was that reported recentlyE3 with a minor modification (use of RPMI instead of AIM-V culture medium to increase infection efficiency by reducing the concentration of serum albumin). Following collection during bronchoscopy, explants were rested overnight in pairs in 24-well culture dishes containing 500mL of RPMI supplemented with glutamine and penicillin/streptomycin in a humidified tissue culture incubator at 37°C, 5% CO2. They were then cultured in RPMI medium supplemented with glutamine alone (RPMI-G) and treated with either 100nM FP or carrier control (0.1% v/v dimethyl sulfoxide) for 2 hours before adding log7.0 plaque-forming units (pfu) of virus or the equivalent volume of virus diluent (0.4% w/v sucrose in 0.5mM HEPES buffer, pH 7.4) in the presence of carrier/FP and incubated for 2 hours. Explants were then washed 3 times with basal RPMI medium to remove excess virus and incubated for a further 22 hours in RPMI-G containing carrier/FP, after which conditioned media were centrifuged (400g) to remove cellular material and stored at -80°C. Samples for plaque assay were stored in 40% (w/v) sucrose-50mM HEPES buffer (pH 7.4).

Flow cytometric analysis of infection and cell activation markers

Post-infection, tissue pieces were digested in RPMI containing 1mg/mL collagenase I for 60 minutes with agitation. The dispersed cells were then filtered through 100mm filters to remove tissue debris and resuspended in 100mL of fluorescence-activated cell sorting buffer (0.5% BSA, 2 mM EDTA in Dulbecco’s PBS) containing 2mg/mL human IgG for 10 minutes on ice. Cells were then stained for leukocytes (the pan-leukocyte marker CD45 conjugated to PECF594), epithelial cells (the epithelial cell marker CD326 conjugated to PerCPCy5.5), T lymphocytes (the T-lymphocyte specific marker CD3 conjugated to phycoerythrin-Cy7), and MHC class II (HLA-DR conjugated to antigen-presenting cell-Cy7) using mAbs directed against each extracellular marker, or appropriate isotype control antibodies conjugated to the relevant fluorophores. Cells were then fixed and permeabilized using proprietary reagents (BD Fix/perm kit, BD Biosciences, Oxford, United Kingdom [UK]), and cells infected with A/H3N2/Wisconsin/67/2005 virus were detected using fluorescein isothiocyanate–conjugated monoclonal anti-influenza nucleoprotein antibody (AbCAM, Cambridge, UK). Flow cytometry of the stained cells was performed using a FACSAria (BD) with appropriate filters and settings. Appropriate isotype and fluorescence-matched antibodies were added to separate samples to aid gating of cell populations (Table E2). Epithelial cells were identified by following a gating strategy modified from previous reports,E3 on the basis of size, and then excluding the leukocyte marker CD45, and including positive staining for the epithelial cell marker CD326 (Fig E1, A). Viral infection in epithelial cells was gated against epithelial cells from uninfected tissue stained with the mAb against viral nucleoprotein (Fig E1, B). T lymphocytes were identified from the CD45-positive population by expression of the T-lymphocyte marker CD3. The applied viral concentration was retitrated in these altered conditions using the previously described gating strategy, to ensure that epithelial cell infection versus viral dose was linear 24 hours post-infection (Fig E1, C), resulting in an optimized final dose of log7.0 pfu. For quantification of cell surface HLA-DR ligand expression on T lymphocytes and epithelial cells, specific mean fluorescence intensity (sMFI) was calculated by subtracting the MFI of the cell population stained with isotype antibody from the MFI of the cell population stained with specific antibody (Fig E1, D). Fold-induction of the expression of HLA-DR on the surface of T lymphocytes was used as a marker of T-cell activation.E4 Data were analyzed using FACS Diva software v5.0.3 (BD).

Viral shedding plaque assay

Cell-free tissue conditioned medium samples containing 40% (w/v) sucrose and 50mM HEPES buffer (pH 7.4) were stored at -80°C until use. Samples were then thawed and immediately prepared by serial dilution in infection buffer (Dulbecco modified Eagle medium containing L-glutamine, sodium pyruvate, penicillin streptomycin, and nonessential amino acids). Diluted media samples were then applied to 90% confluent monolayer cultures of MDCK cells with agitation for 1 hour at 37°C, 5% CO2. Media were then removed and the cultures overlayed with cellulose/methylcellulose biopolymer (Sigma-Aldrich, Poole, Dorset, UK) prepared in minimal essential medium (MEM), supplemented with sodium bicarbonate, HEPES, BSA, and diethylaminoethyl-dextran hydrochloride. Overlay was also supplemented with TRTPCK trypsin (Worthington Biochemical Corp, Reading, Berks, UK) at a final concentration of 0.25ng/mL. Cultures were incubated for a further 48 hours and then the overlays removed, and the monolayers stained with crystal violet to visualize the plaques. Plaques were manually counted and the viral titer (pfu/mL) was then adjusted for the dilution factor and expressed as the number of pfu per milliliter.

Measurement of inflammatory mediators

A bespoke 8-plex immunoassay to quantify a range of cytokines/chemokines (IFN-, IL-6, IL-8, CXCL-10/interferon gamma-induced protein 10 (IP-10), monocyte chemoattractant protein 1, macrophage inflammatory protein 1, TNF-α, and IL-1) was purchased from Mesoscale Discoveries (Rockville, Md) and performed according to the manufacturers’ instructions. Briefly, conditioned media samples were diluted 1:1 in buffer containing carrier protein and applied to microtiter plates precoated with antibodies directed against the 8 mediators to be measured. A dilution series of mediators in the culture medium diluent was also applied to the plate to construct standard curves of mediator concentration. A detection antibody prediluted in blocking buffer was applied, and then the signal analyzed on a SECTOR 2400 MSD plate reader. Data were analyzed using MSD discovery workbench software. Cytokine quantities were interpolated from standard curves generated using mixtures of recombinant proteins. Baseline secretion from noninfected untreated controls was subtracted from infected and/or treated wells. Values were then adjusted for wet tissue weight (mg) to give the concentration of mediator per milligram per milliliter culture medium.